The issue about "A truncation error was encountered trying to shrink CHAR '65535' to length 1" should now be fixed with latest JPOX. JPOX CVS now uses default "jdbc-type" as per the most recent JDO spec. In particular for this error, when the java type is Character/char the jdbc-type chosen will be CHAR(1) (where the RDBMS supports it, which Derby does).
Test errors/failures previously were 61 for application-id, 60 for datastore-id. Now we have 51 for application-id, 46 for datastore-id. and moreover, of the remaining errors there are 15 for application-id, and 11 for datastore-id giving "SQL Exception: Invalid character string format for type INTEGER". This occurs where the schema has been defined with an INTEGER column to represent a Character/char(/String) java type (maybe to allow for previous JPOX behaviour) and implies that the schema needs some updates to match the default jdbc-type expected by the JDO2 spec (and now also by JPOX). Schema where column is INTEGER but should be CHAR(1) FieldsOfCharacter : table FIELDSOFCHARACTER has INTEGER columns FieldsOfPrimitivechar : table FIELDSOFPRIMITIVECHAR has INTEGER columns Schema where column is INTEGER but should be VARCHAR(...) HashMapStringKeyCollections, HashMapStringValueCollections, MapStringKeyCollections, MapStringValueCollections, TreeMapStringKeyCollections, TreeMapStringValueCollections : join tables have the key as INTEGER instead of VARCHAR There may be others of the same nature that I've not listed, just wanted to give a head start on fixing the schema. Also, this change may affect other JIRA issues, but sadly "http://issues.apache.org"; is about as unresponsive a server as I've ever come across, so we'll have to leave that for others to work out ;-) -- Andy
